RFQ: Innovative Credit Scoring (ICS) Partner

Roles and Responsibilities

Women’s World Banking, a leading, global non-profit organization, is building a world where low-income women, their families, businesses, communities, and economies thrive. We work on the ground with financial services providers to demonstrate the benefit of investing in women as customers and as leaders. We conduct research to build the evidence base for women’s financial inclusion. We drive systems-level change by advocating to policy makers in developing countries to create policies that support women’s financial inclusion, increase gender diversity, and build women’s leadership. 

Across Southeast Asia, and Indonesia in particular, many women-owned micro and small enterprises face barriers to accessing formal credit due to limited collateral, low digital footprints, and minimal or no traditional credit histories. 

To address this, our project aims to pilot and validate innovative credit scoring models that leverage alternative datasets (including behavioral, transactional, telco, and psychometric data) to complement financial institutions’ internal credit assessment processes. 

We are seeking an experienced ICS partner with demonstrated capabilities in developing and deploying machine learning–driven credit scoring models and integrating alternative data sources relevant to the women’s segment. 

The selected Innovative Credit Scoring (ICS) provider will work collaboratively with Women’s World Banking (WWB) and financial service providers (FSPs) partner to explore and pilot innovative credit scoring approaches that can help expand access to productive credit for low-income and thin-file customers, including recipients of government-to-person (G2P) payments such as PKH. 

The selected ICS provider will be responsible for but not limited to: 

1. Identify, analyze and map data sources 

• Conduct a rapid assessment of relevant data sources that could help assess the creditworthiness for women customers, including but not limited to: 

• Internal FSP data (such as transaction, digital payment such as QRIS, e-statement, existing credit report or repayment data) 

• External data (such as telco, mobile device, tax, multi-platform, utilities, e-commerce, BPJS, social media) 

• Evaluate data relevance, availability, quality and integration feasibility to the partner’s system (if applicable). 

• Analyze and provide insights of any data that is relevant to the project including credit portfolio data from the FSP partner, targeted customers data/information, business owner/merchant and their transactions data, etc. 

• Provide recommendations on alternative data that relevant to predict income or repayment capacity and behavior of ultra micro and micro businesses, particularly women entrepreneurs. 

2. Provide access to alternative datasets and data integration support 

• Provide or facilitate access to a wide range of high-quality, alternative datasets that are commonly used in innovative credit scoring for underserved segments. 

• Support basic data preparation including data cleaning and pre-processing 

• Work with FSP partners to combine alternative data with existing internal data through simple data-sharing arrangements or API-based integration, depending on technical feasibility upon assessment. 

• Follow good data governance and consent practices appropriate to Indonesian context. 

3. Improve overall credit assessment process 

• Provide expertise and knowledge on overall risk management practices, particularly on inclusive credit risk assessment and alternative scoring approaches for productive credit products. 

• Collaborate with the WWB team and experts to improve credit assessment processes of partner FSPs 

4. Develop and test the credit scoring model with different approaches 

• Develop and test robust credit scoring models with alternative data, using appropriate analytical approaches, such as logistic regression, decision trees, machine learning or hybrid methods, where relevant. The credit scoring model should be able to, but not be limited to, predict income or repayment capacity and behavior to mitigate credit risks. 

• Conduct data analysis to identify key predictors and patterns relevant to women’s creditworthiness 

• Collaborate closely with the partner financial institution’s risk team to co-design and train credit scoring models tailored to thin-filed women customers 

• Conduct back-testing and validation exercises using available or pilot data 

5. Support pilot implementation, monitoring and evaluation 

• Provide technical expertise during the pilot and monitoring phase 

• Ensure ongoing data quality and performance monitoring 

• Share recommendation for adjustment based on pilot outcomes 

Desired Experience and Expertise 

• Registered or licensed as credit scoring model firm that can use alternative data in Indonesia 

• Proven track record in developing and deploying alternative data-based credit scoring models for productive credits for micro and small businesses. 

• Familiarity with Indonesian market context including its data infrastructure landscape relevant for credit scoring model with alternative data. 

• Have access to or established partnerships with wide-range data providers (e.g., telcos, payment providers, e-commerce platforms, BPJS, Tax, Utilities). 

• Experience working with financial institutions (Banks and/or Non-Banks) in Indonesian markets, with local presence or operational footprint in Indonesia. 

• Demonstrated expertise in data science, machine learning, and risk analytics 

• Able to work within the FSP partner working environment 

The following experience is preferred but not required and will be considered an added advantage. 

• Familiarity with inclusive finance, gender-lens strategy, or women-focused financial products preferred 

• Understanding of the micro and ultra-micro segment context in Indonesia. 

• Additional experience working with low-income population, including G2P recipients or social assistance programs (such as PKH in Indonesia or similar programs in other countries).